The present invention relates to network troubleshooting, and more specifically, to a network element diagnostic evaluation.
Network administrators or technicians typically need to diagnose underperformance in a network. Conventional approaches require an administrator to perform diagnosis locally at the network element. Debugging and diagnosing various network and switching gears (referred to hereinafter, as “network elements”) is typically performed by collecting large memory dumps and/or large amounts of packet captures and analyzing the data for performance or errors. Analysis of memory dumps and packet captures may require huge computing power and large amounts of memory, typically performed by a local machine. In order to perform diagnosis and debugging, the local machine is connected to the network element so that software resident on the local machine can obtain the memory dump or packet capture and perform the intensive computing during analysis. Thus, under conventional approaches, a local machine with sufficient computing power should either be permanently connected to the network element or should be shuttled around to each network element for connection and diagnosis. As may be appreciated, this can cause unnecessary labor and expense.